docs: Added documentation for the improved code review suite

Agents.md

@@ -877,7 +877,12 @@ Coyote comes packaged with some useful built-in agents:
    changes consumer-facing surface), and finishes with ONE draft PR, CI checks watched to green,
    plus follow-up task files for every manual action the code can't do for itself. Ends with a
    `whetstone` continuous-improvement pass when the run addressed review feedback from outside its
-   own lanes. Supports opt-in
+   own lanes. An `escalation_policy` variable governs decision points: `autonomous` answers them (its
+   own and its Sisyphus children's escalations) from standing rulings in
+   `<config_dir>/share/standing-rulings.md` or a clearly recommended option, amends plan text before any
+   conformance re-run, and batches every self-made call into a `Decisions taken autonomously` digest in
+   the final report and PR body: secrets, money movement, and destructive actions always escalate to
+   you. Supports opt-in
    worktree-based parallel task execution. Plans declare a quality bar: `rigor` (poc/prototype/production)
    plus the `surfaces` being built (rest-api, cli, library, worker, iac, db-migration, frontend, ci-cd, or
    `other:<label>` via a design-time `librarian` lookup), which flows into every downstream reviewer, with
@@ -975,7 +980,12 @@ Coyote comes packaged with some useful built-in agents:
    surface, and `probe` for black-box usage-pattern verification when the change touches
    consumer-facing surface. After a run that addressed review feedback from outside its own lanes (humans,
    or a teammate's independent review agent), it spawns `whetstone` once to fold the review misses back into
-   the suite as user-approved improvements. Also supports plan-driven workflows: authoring phased implementation plans (via the `plan-authoring` 
+   the suite as user-approved improvements.
+   An `escalation_policy` variable governs mid-run decisions: `autonomous` resolves them from standing
+   rulings (`<config_dir>/share/standing-rulings.md`, recorded via the `record-ruling` macro) or a
+   clearly recommended option, amends plan/criteria text before any re-review, and reports every
+   self-made call in a `Decisions taken autonomously` digest: secrets, money movement, and destructive
+   actions always escalate. Also supports plan-driven workflows: authoring phased implementation plans (via the `plan-authoring` 
    skill), having `oracle` review them, and executing them one reviewed step at a time. Handles review findings
    quality-bar-aware: the plan's `rigor` calibrates which severities block (🔴 always does) and maps to a default
    `security_posture`; below-threshold `[convention]` findings defer into follow-ups instead of blocking; and a

Macros.md

@@ -212,6 +212,9 @@ looking for more examples.
 * `generate-commit-message` - Generate a Git commit message based on the staged changes in the current directory
 * `record-miss` - Record a code review suite miss into the review miss ledger at `<config_dir>/share/review-misses.md` so that it is not missed
   on future code review suite runs.
+* `record-ruling` - Record a standing ruling (a pre-delegated decision) into `<config_dir>/share/standing-rulings.md`. Agents running
+  with `escalation_policy: autonomous` apply a matching ruling instead of escalating to you, and report every application in the
+  run's `Decisions taken autonomously` digest. Hard triggers (secrets, money movement, destructive actions) can never be delegated.
 
 Built-in macros are written to your macros directory on first run and never overwritten afterward, so your edits are
 preserved across Coyote updates. To discard local changes and reinstall the built-in macros from the current Coyote build,